Optimizing passengers and goods flows reduces the costs and ecological footprint of transportation systems, as well as congestion in cities. Since passengers and goods do not have the same nature (a person is active and a good is passive), modeling each flow separately seems natural. But local authorities need to consider passenger and freight transport together as a single logistic system. In this paper, we assess the efficiency of an existing transportation system where the spare capacity of public transport is used to distribute goods toward the city core. We propose two optimization approaches. The first one considers the problem as a multidepot VRPTW followed by an assignment problem. The second one solves the whole model as a pickup and delivery problem with transfers (PDPT). These approaches are evaluated on data sets generated following a field study in the medium-sized city of La Rochelle in France.